[Insight-users] watershed segmentation without RGB?

[Laurent Mundeleer]

Thanks a lot for the long answer, I'll try it and let you know what I've 
got :)

regards,

Laurent


Miller, James V (Research) wrote:

>Watershed can be run on grayscale images.  The example
>(WatershedSegmentation1.cxx) is set up to process RGB pixels. But you can
>easily run watershed on a grayscale image.
>
>You'll want to create a similar program to WatershedSegmentation1.cxx that
>reads a scalar image, produces a gradient magnitude image
>(GradientMagnitudeImageFilter or RecursiveGradientMagnitudeImageFilter), and
>runs watershed on the gradient magnitude image.
>
>So you just need to change the pixel type, delete the word "Vector" from any
>class type in example, and change the CastType to a CastImageFilter that
>converts a scalar image into an image of floating point pixels (alternative,
>you can remove the cast filter altogether and have the reader return an
>image of floating point pixels.  Here is a quick stab at it (haven't tried
>to compile it....)
>
>#ifdef _MSC_VER
>#pragma warning ( disable : 4786 )
>#endif
>
>#include <iostream>
>
>#include "itkGradientAnisotropicDiffusionImageFilter.h"
>#include "itkGradientMagnitudeImageFilter.h"
>#include "itkWatershedImageFilter.h"
>
>#include "itkImageFileReader.h"
>#include "itkImageFileWriter.h"
>#include "itkUnaryFunctorImageFilter.h"
>#include "itkScalarToRGBPixelFunctor.h"
>
>int main( int argc, char *argv[] )
>{
>  if (argc < 8 )
>    {
>    std::cerr << "Missing Parameters " << std::endl;
>    std::cerr << "Usage: " << argv[0];
>    std::cerr << " inputImage outputImage conductanceTerm
>diffusionIterations lowerThreshold outputScaleLevel gradientMode " <<
>std::endl;
>    return 1;
>    }
>  
>  typedef itk::RGBPixel<unsigned char>   RGBPixelType;
>  typedef itk::Image<RGBPixelType, 2>    RGBImageType;
>  typedef itk::Image<unsigned long, 2>   LabeledImageType;
>  typedef itk::Image<float, 2>           ScalarImageType;
>
>  typedef itk::ImageFileReader<ScalarImageType> FileReaderType;
>  typedef itk::GradientAnisotropicDiffusionImageFilter<ScalarImageType,
>    ScalarImageType>  DiffusionFilterType;
>  typedef itk::GradientMagnitudeImageFilter<ScalarImageType>
>    ScalarMagnitudeFilterType; 
>  typedef itk::WatershedImageFilter<ScalarImageType> WatershedFilterType;
>  // Software Guide : EndCodeSnippet
>
>  typedef itk::ImageFileWriter<RGBImageType> FileWriterType;
>
>  FileReaderType::Pointer reader = FileReaderType::New();
>  reader->SetFileName(argv[1]);
>  
>  DiffusionFilterType::Pointer diffusion = DiffusionFilterType::New();
>  diffusion->SetNumberOfIterations( atoi(argv[4]) );
>  diffusion->SetConductanceParameter( atof(argv[3]) );
>  diffusion->SetTimeStep(0.125);
>
>
>  GradientMagnitudeFilterType::Pointer gradient =
>GradientMagnitudeFilterType::New();
>  gradient->SetUsePrincipleComponents(atoi(argv[7]));
>
>
>  WatershedFilterType::Pointer watershed = WatershedFilterType::New();
>  watershed->SetLevel( atof(argv[6]) );
>  watershed->SetThreshold( atof(argv[5]) );
>
>
>  typedef itk::Functor::ScalarToRGBPixelFunctor<unsigned long>
>    ColorMapFunctorType;
>  typedef itk::UnaryFunctorImageFilter<LabeledImageType,
>    RGBImageType, ColorMapFunctorType> ColorMapFilterType;
>  ColorMapFilterType::Pointer colormapper = ColorMapFilterType::New();
>  
>  FileWriterType::Pointer writer = FileWriterType::New();
>  writer->SetFileName(argv[2]);
>
>  diffusion->SetInput(reader->GetOutput());
>  gradient->SetInput(diffusion->GetOutput());
>  watershed->SetInput(gradient->GetOutput());
>  colormapper->SetInput(watershed->GetOutput());
>  writer->SetInput(colormapper->GetOutput());
>
>  try 
>    {
>    writer->Update();
>    }
>  catch (itk::ExceptionObject &e)
>    {
>    std::cerr << e << std::endl;
>    }
>    
>  return 0;
>}
>
>
>
>
>
>-----Original Message-----
>From: Laurent Mundeleer [mailto:lmundele at ulb.ac.be]
>Sent: Thursday, April 29, 2004 6:10 AM
>To: insight-users at itk.org
>Subject: [Insight-users] watershed segmentation without RGB?
>
>
>Hi all,
>
>I'd like to apply the watershed filter to CT images, is it possible? In 
>the documentation, it 's specified that RGB values are required...
>How can I do?
>
>Thanks in advance
>
>Regards,
>
>Laurent
>
>  
>

-- 
********************************************
Laurent Mundeleer
Université Libre de Bruxelles (ULB)
Service des Systèmes Logiques et Numériques (SLN) CP165/57
50, Av. F.Roosevelt
1050 Bruxelles
Belgium
tel : ++32.2.650.22.97
fax : ++32.2.650.22.98
e-mail : lmundele at ulb.ac.be
********************************************